There is no better marker of the rickety psychiatric infrastructure in the USA than the lack of psychiatric beds. A close second is how those beds are utilized to basically run patients in an out to maximize hospital profits. It seems like I have said it a thousand times on this blog but I will say it again - hospitals make money by getting psychiatric patients out in advance of the diagnosis related group (DRG) time limit. These days that it is about 3-4 days. If management believes that the psychiatrist is not discharging people fast enough - they will turn up the heat on them to do so by using either a designated case manager or somebody who sits in team meetings and reports that psychiatrist to his or her superiors if the patients are not out by a maximum of about 6 days.
There are huge problems with that business approach to psychiatric care. The first is patient complexity. Severe psychiatric disorders place people at risk for significant medical problems and often psychiatric care cannot proceed until those medical problems are stabilized. During my career for example I had terminal cancer patients and patients with uncontrolled diabetes mellitus and hypertension admitted directly to my care because they had a major psychiatric disorder. Substance use disorders complicate at least half of the admissions and psychiatric care typically has to wait until a patient is detoxified from an intoxicant. Very ill patients with schizophrenia and mood disorders who received outpatient treatment cannot be treated and stabilized in 4-6 days. Specific problems like suicide risk and delirium often take many weeks of care. Although brief stays can be useful in the case of event or intoxicant related crises the length of stay on psychiatric units is basically an arbitrary number of days determined by bean counters rather than doctors. They do no reflect clinical reality.
That brings me to the commentary by Sisti, Sinclair, and Sharfstein (1). They lost me then they had me and then they lost me completely. My first criticism is the title "Bedless Psychiatry-Rebuilding Behavioral Health Service Capacity." Ironic that the authors use the managed care buzzword "behavioral health" to suggest that the bed crisis can be addressed by the same carpetbaggers that designed the current system. I can appreciate a political turn of a phrase as well as the next rhetorician, but in the case it falls very flat. The only way to address the bed crisis and the destruction of the mental health care infrastructure in this country is to get rid of managed care and all of their buzzwords. There is no way that companies paid well for rationing care and kicking unstable people out of psychiatric hospitals are going to solve that problem.
From there the authors do an adequate job of describing the problem of a sharp drop in bed capacity in addition to the absurdly short lengths of stay. They depend on data that may have another agenda. In a recent post in this blog, I looked at the drop in state hospitals beds in Minnesota and the Medical Directors commentary on why that will never be reversed. The same organization that authored the report used by the authors to describe the drop in beds (National Association of State Mental Health Program Directors (NASMHPD) is on record stating that "Building more inpatient bed capacity to meet demand is unsustainable". State Mental Health Program Directors are all accountable to state politicians and generally run state mental health programs like managed care companies do. They ration services and limit access to treatment. It is cost effective from their perspective to leave large blocks of people untreated. Better yet put them in jail and give them a baloney sandwich everyday instead of the à la carte fare that medical and surgical patients have come to expect in customer satisfaction based hospitals. This conflict of interest and lack of interest in looking at whether bed capacities are too low is a bias that any reader of the report should be aware of. They also consider OECD data and suggest that psychiatric bed capacity in the USA is 4th from the lowest bed capacity in the countries studied.
They go on to discuss the "types" of beds and suggest that the notion that bed capacity may be too abstract. They favor bed descriptions based on the function of the unit that they reside on - forensic, acute care, intermediate, and long term care. They discuss beds in the grey zones between corrections and mental health. For example in my discussion of the Minnesota situation, I did not include beds operated by the Minnesota Sex Offender Program (MSOP). That program houses 726 clients at two large facilities or about three times the state bed capacity for all of the committed patients with mental illness in the state. In a bizarre end run around psychiatry, sex offenders in the state are essentially granted mental illness status. This occurs in order to allow the state to indefinitely commit them. MSOP clients are essentially never discharged while committed patients back up and crowd local hospital psychiatric units and shut them to new admissions while they are waiting to be transferred into the state hospital system. The argument about no new beds at the state level does not apply to sex offenders.
The authors close by saying the concept of a psychiatric "bed" may need to be "jettisoned" in order to more accurately address the needs of patients and system capacity. They end with the idea that "targeted payment reforms" are necessary to increase psychiatric bed capacity. I think that they have it wrong on both accounts. We have had 30 years of "incentives" that really are not incentives. The DRG payment itself was allegedly a payment for what was the average amount of care for a particular diagnosis. Instead, it became a way that managed care companies could game the system while they rationed care. It may not be as easy to determine (another bean counter bias) - but looking at the flow though systems and where services are short is a better idea. Classic examples are outpatient psychiatrists who are not able to refer one of their outpatients to an inpatient unit in the same system for purposes of detox, electroconvulsive therapy, or stabilization. Whenever that happens it should be taken as a sign that health plan needs to improve their bed capacity.
Bed quality is as least as important as bed inventory. Beds are worth less if there are problems with the physical structure or staffing problems. Beds are worth less if a therapeutic environment cannot be maintained. Beds that can contain aggressive behavior are generally at a premium because fewer people can work in that setting. In every state there are only a few psychiatric units that will address aggression as a psychiatric problem. Specialty units to treat depression, bipolar disorder, schizophrenia, substance use in addition to mental illnesses, or medically ill psychiatric patients are rare. There appears to be no interest in either the quality or specialty side. DRG payments create an incentive to get people out as soon as possible and provide the lowest level of quality.
A very basic comparison with any systems of high quality beds that address the medical problem with state of the art care is instructive. Any middle aged person in the US who presents to the emergency department with chest pain who has cardiac risk factors will be admitted to a telemetry unit, get the necessary blood testing, and (if all of those tests are negative) will probably get an echocardiogram and cardiac stress test before they leave the next day. That same person presenting to the emergency department with hallucinations or mania or severe depression or delusions will only be admitted unless they are determined to be "dangerous". The standard definition of dangerousness being "imminent risk of harm to yourself or others." Dangerousness is the managed care approach to psychiatric hospitalizations. It contaminates emergency assessment and it contaminates what happens on the inpatient side. When the overriding treatment dimension is dangerousness - inpatient units become holding tanks where nothing therapeutic occurs. Patients sit around and look at one another all day long waiting for someone to proclaim that they are no longer dangerous - so they can be discharged. Beds that operate under this punitive model should probably not be counted.
The authors' commentary seems to continue the same policy wonk approach that has contaminated practically all medical journals - basically a number of administrators sitting around and speculating. Unfortunately we know that a lot of bad ideas get started this way. We also know that hypotheticals and incentives have have been the order of the day for a generation and that very process knocked out bed capacity and led to all of this low quality care.
To improve the bed capacity it will take a psychiatrist who is aware of the problems and how they can be addressed in each state. Being on the ground as the inpatient beds and any quality they had were rationed away would be a plus. Knowing how to build increased capacity and quality is the best possible approach.
George Dawson, MD, DFAPA
References:
1: Sisti DA, Sinclair EA, Sharfstein SS. Bedless Psychiatry—Rebuilding Behavioral Health Service Capacity. JAMA Psychiatry. Published online March 14, 2018. doi:10.1001/jamapsychiatry.2018.0219
Graphics Credit:
The above picture of an abandoned state hospital bed is downloaded from Shutterstock per their standard licensing agreement.
The above picture of an abandoned state hospital bed is downloaded from Shutterstock per their standard licensing agreement.